zlacker

I was working on a similar IO problem with wearables a while ago (though by the sounds of things, far less seriously than you are), and I had the idea that maybe that band/strap could function as an on-off switch, so when you undo the band (which you do when taking it off), it turns the device off, and vice versa. Could be something you could try too

>>bodge5+(OP)
I'm putting capactive sensors in my wearables to turn them on when in contact with a human.

I strongly believe that the class of widget I'm building should stay firmly out of the user's way. The point is to forget it's there. So, as simple IO as possible.

>>bodge5+(OP)
This is a fun material you could use to detect if a band was stretched or not, silver coated elastic [0], near 0 ohm resistance when loose, resistance increases when stretched. I built a voltage divider with a patch of it when I was experimenting with fabric input devices, mostly just noise makers, but you can see how responsive it is [1]

[0] https://lessemf.com/product/stretch-conductive-fabric/

[1] https://youtu.be/Xjo4w4OiBS8

>>bodge5+(OP)
Polar H10 basically works like that. It's great until you are suddenly unable to connect to it via bluetooth, and have to blindly debug the reason. Is the device even on? Battery low? Broken? App issues? Who the fuck knows.

>>bodge5+(OP)
We've taken a similar approach, I won't divulge exactly what we've done yet, but our goal was to not have a button that would have to be placed somewhere that it wasn't going to get pressed in the middle of the night, would have to be large enough to be usable, adding thickness to the headband at some point.

>>_Alger+kf1
Yeah, you hit on a very important point. There are ways to handle this via lighting on the device (as far as, connected, battery, etc), and I think the app being clear about the state, beyond just "connected/not connected" is important.

We've had very few issues with BLE, but it is one of the things I'm most worried about as it absolutely kills the experience.